[openbox] keyboard shortcuts for the root menu

[Clay Barnes]

On 19:45 Thu 19 Jul     , Dana Jansens wrote:
> On 7/19/07, Clay Barnes <clay at hci-matters.com> wrote:
> >On 12:50 Thu 19 Jul     , Dana Jansens wrote:
> >> On 7/8/07, Jonas J Linde <jonas at init.se> wrote:
> >> >And Jonas J Linde spoke unto the world. And said:
> >> >>And Jonas J Linde spoke unto the world. And said:
> >> >>>And Dana Jansens spoke unto the world. And said:
> >> >>>>On 6/29/07, Jonas J Linde <jonas at init.se> wrote:
> >> >>>>If you want to go ahead and make it do all the stuff I've talked 
> >about
> >> >>>>here (exclusively use _, without the option) against the 3.4-working
> >> >>>>branch in SVN that would be cool, or I can do it when I get to it as
> >> >>>>well.
> >> >>>
> >> >>>OK, I'll start working on it.
> >> >>
> >> >>Here's a new patch - this one against the 3.4-working branch. Skipping
> >> >>the shortcut color bit made the patch a lot simpler although switching
> >> >>from & to _ made it seven times larger.
> >> >>
> >> >>I'll see if I can find a better way to use color instead of underline
> >> >>and get back to you when I do.
> >> >
> >> >Hi again,
> >> >
> >> >So, here's the patch to allow menu.items.shortcut.color, (and optionally
> >> >menu.items.active.shortcut.color, menu.items.disabled.shortcut.color and
> >> >menu.items.active.disabled.shortcut.color) in the themerc file. I
> >> >couldn't figure out a good way to remove an attribute from the Pango
> >> >layout so it got a bit complicated with five layout attributes out of
> >> >which only one is used at a time. I'm not convinced that this is the
> >> >Correct Way to do it but at least it's a way.
> >> >
> >> >As I hinted previously, if menu.items.shortcut.color isn't set,
> >> >underlining will be used instead so as not to break old themes.
> >> >
> >> >Would this be OK for inclusion in the next release?
> >>
> >> So I've been thinking about it.  I have a branch in my repository
> >> right now with this functionality in it.  I changed the theme element
> >> names up, and did the code quite differently - the patch provided
> >> broke all the render lib's (shotty) encapsulation.  But I'm really
> >> hesitant to add this still, because for 99.9% of users, when they turn
> >> on this option, nothing will happen, and they will consider it broken,
> >> and file a bug report.  Maybe just using the opposite of the current
> >> color would be better, like an xor operation does.  I will continue to
> >> think about it, but it might not show up in the next release.
> >>
> >> Dana
> >>
> >
> >From an HCI standpoint, this option is a dangerous one.  While
> >well-chosen colors won't do much harm, theme designers and tweakers
> >would have to be fairly well-versed in color theory to be able to make
> >that selection well.  For example, blue is focused in a different way
> >than other colors, meaning a blue letter with the rest being black
> >letters would render the blue letter surprisingly hard to read (unless
> >shades and tones to the two colors are chosen very carefully).
> >
> >Additionally, many designers will be tempted to choose colors that are
> >poorly (if at all) differentiable by people of limited color
> >perception (color-blind, colloquially).
> >
> >I think we're much better off with the traditional underlining,
> >despite its lower visibility in some cases.
> 
> Yeah, I agree with all of that, it's definitely an important
> consideration.  The only point against it is that it would be optional
> (and default off).  And color contrast is already a big issue in menus
> with many themes.
> 

My concern, then, is that it seems rather against OB design
principals to add a feature that would be rarely used as well as have
serious, recognized design problems.  I hope, if it does get included,
that it would be an option at build-time.

-- 
Clay Barnes

Website:
http://www.hci-matters.com

GPG Public Key:  
http://www.hci-matters.com/keys/claybarnes_public_key_until20080718.gpg
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://icculus.org/pipermail/openbox/attachments/20070719/dec64989/attachment.pgp>